Every few years, the market forms a wave of collective enthusiasm around some new area of technology. Industrial AI, new energy systems, advanced materials, the low-altitude economy and synthetic biology are several fields currently in that phase. What these fields share is that the technical path has not yet converged, business models are still being tested through trial and error, and regulatory frameworks typically lag industry practice. Our approach to fields like these is not to bet in advance on which direction will win, but to build research capability first and only then decide whether capital should enter, and at what scale. This page sets out that method itself, not a forecast of the prospects of any specific frontier field.
A watch list, not an investment list
Industrial AI, energy systems, advanced materials, the low-altitude economy and synthetic biology currently sit in our framework as themes under continuous observation, not as an established strategy direction with defined investment criteria already in place. This distinction matters: observation means the research team invests time in understanding the technical path, industry-chain structure and potential demand sources, but it does not necessarily mean capital has entered or is about to. Conflating observation with commitment is a common starting point for misjudgment among capital operating in emerging fields — following a field's developments closely, and concluding that the field has reached conditions mature enough to invest in, are two separate decisions.
Entry thresholds: three conditions that must be cleared
For an observed theme to convert into an actual capital commitment, the following conditions need to be satisfied together, not satisfied one at a time.
Real paying customers, not pilot partnerships
Technology demonstrations and pilot partnerships are comparatively easy to arrange; the real test is whether customers exist who are willing to keep paying and have folded the product into daily operations. However many pilots accumulate, if none converts into a formal purchase, the underlying market demand has not yet been genuinely established.
Unit economics visibly clear
The relationship between customer acquisition cost, delivery cost and customer lifetime value needs to be observable and verifiable, not resting on an assumption about future scale economics. The narrative that costs will fall once scale arrives is common, but before that scale actually materializes, the narrative cannot serve as a basis for investment.
A largely clear regulatory path
Fields such as the low-altitude economy and synthetic biology depend especially on regulatory maturity; only once key rules — airworthiness, market access or safety standards — are largely settled does a commercialization timetable become predictable enough for capital to form a reasonable view of the return horizon.
How research precedes capital in practice
For themes on the watch list, the research team's working method differs from sector coverage under an already-established strategy: it relies more on direct exchanges with participants up and down the industry chain, ongoing mapping of where technical paths diverge, and close tracking of regulatory developments, rather than on public market information or third-party reports. This kind of research often produces no investment decision for years at a stretch. Its value lies in ensuring that once conditions genuinely mature, the team already has enough industry understanding to make a high-quality judgment the moment a window opens, rather than starting to learn an unfamiliar field from scratch.
The path from observation to capital deployment
- Theme added to watch list
- Industry-chain mapping and practitioner interviews
- Tracking technical paths and regulatory developments
- Ongoing testing against entry thresholds
- Thresholds unmet: observation extended
- Thresholds met: small exploratory deployment
- Position sized up after validation
Theme added to watch list: 1; Industry-chain mapping and practitioner interviews: 2; Tracking technical paths and regulatory developments: 3; Ongoing testing against entry thresholds: 4; Thresholds unmet: observation extended: 5; Thresholds met: small exploratory deployment: 6; Position sized up after validation: 7
Sizing positions under high uncertainty
Even once entry thresholds are met, uncertainty in an emerging field usually remains considerably higher than in a mature industry, which is why the initial position should stay restrained. We prefer to make a first entry at a small scale, sized so the position could be fully lost without disproportionate consequence, and then decide whether to add based on predefined validation checkpoints — customer renewal rates, or the actual trajectory of unit-cost decline, for example — rather than committing a large amount of capital in one step before conviction has been tested. This staged approach to sizing is, in essence, a way of converting uncertainty into a manageable, stage-released risk exposure, rather than an attempt to eliminate the uncertainty itself.
Two preconditions that support this method
Capital with no fixed horizon
The path from opening research to conditions maturing in an emerging field often takes years; only capital not bound by a fixed exit timetable can genuinely wait rather than force an outcome.
Willingness to bear research cost without expecting immediate return
Years of sector tracking are themselves a cost, and are likely, more often than not, to convert into no specific investment in the near term. This commitment needs to be understood as long-term capability-building, not output measurable on any immediate basis.
The Havrion Capital Perspective
Our stance toward emerging industries is restraint rather than passive observation: we keep committing research resources to understand a field, while explicitly declining to deploy capital at meaningful scale until the three thresholds — customers, unit economics and regulatory path — are satisfied together. This stance sometimes means that after watching a theme discussed enthusiastically in the market for several years, we remain in the observation phase. That is not hesitation; it is what the method itself requires.
We believe that in a genuinely unfamiliar field, method is more reliable than prediction. No one can reliably judge in advance which frontier technology will ultimately outperform the alternatives, but a clear set of entry thresholds and a staged sizing discipline lets us move quickly, with sufficient industry understanding, once conditions mature — and lets us avoid taking on unnecessary risk prematurely, driven by market sentiment, when conditions have not. That method is itself the principal asset we bring to emerging industries.
Related Portfolio Companies
Related Insights
Where AI Adoption Is Real in the Industrial Economy
Most industrial AI deployments remain pilots. A smaller set has moved into repeatable, cost-accountable operation. What separates the two has less to do with model sophistication than with data access, deployment capability and ownership of the workflow — the three factors that determine whether a vendor compounds or stalls at the next budget review.
7 min readTreating the Energy Transition as an Industrial Buildout
The energy transition is often discussed as a policy narrative, but the investable core of it is a decade-plus industrial buildout — distributed generation, storage attach, industrial electrification and grid-adjacent services, each advancing on its own timeline. This piece sets out how to break those threads into observable indicators, and how to tell an engineering-led business apart from one that depends on a subsidy for its survival.
7 min read