A commercialisation engagement typically begins with a version of the same question: what should the technology actually be, and for whom?
The question is asked of the commercial advisor. The answer depends on information only the engineering team has. The engineering team is solving a different problem that week. By the time the two conversations reconverge, a month has passed, a decision has been made on one side that constrains the other, and the venture has quietly drifted into a position no one chose.
This is not dysfunction. It is the default operating model of most deeptech advisory relationships. And it is where ventures lose the most value.
The integration problem, named
In deeptech, engineering and commercial decisions are the same decision, applied in different registers. The choice of beachhead market determines what the product has to do. The choice of business model determines where the venture should own value and where it should partner — which in turn shapes what the engineering team has to build, what can be licensed, what can be outsourced, and what must be proprietary.
Treating these as sequential — define the market, then scope the engineering, then revisit the business model — is the standard failure pattern. The decisions are coupled. Making them sequentially means each one is made with incomplete information about the others.
The alternative is to make the decisions together, in the same room, with both capabilities represented.
What the split actually costs
There are three costs, and they are substantial.
The first is time. Every decision made on one side that has to be revisited on the other is a round trip of weeks or months. Three or four such round trips is a year of runway.
The second is optionality. Engineering choices made without commercial context frequently close off applications the venture has not yet discovered it should pursue. Commercial choices made without engineering context frequently commit to markets the technology cannot actually serve at scale. Both kinds of mistake are harder to undo than to avoid.
The third is narrative. By the time the venture presents to investors, the engineering story and the commercial story have been built on different timelines by different people. The gaps show. Sophisticated investors read those gaps as an indicator of how the venture will be run post-investment — and price accordingly.
What integration actually looks like
Integration is not a meeting format. It is an operating model.
In an integrated engagement, the same two or three people are responsible for both the engineering roadmap and the business model. They attend customer conversations together. They review the commercial screen and the technical roadmap in the same session. When a commercial decision is proposed, the engineering implications are stress-tested on the spot; when an engineering decision is proposed, the commercial implications are stress-tested on the spot.
The point is not that everyone in the room has dual expertise. The point is that the two expertises are applied to the same problem at the same time, with the same shared context.
This is uncomfortable for the standard advisory model. It requires a firm that has both capabilities in house rather than stitched together across partners. It also requires the people with those capabilities to actually trust each other’s judgement enough to make joint calls rather than staging them through a client.
The artefacts of integration
A venture that has worked this way produces different artefacts than one that has not.
The engineering roadmap reads as a product roadmap. Milestones are tied to customer-observable capabilities, not to academic validation. Cost-down and manufacturability appear as first-class concerns, not as post-hoc additions.
The business model reflects actual engineering constraints. Where the venture owns value and where it partners is a decision, not a default — and the decision reflects where differentiation is real and defensible, not where ambition is highest. Our engagement with eThermflex repositioned the venture from an integrated product company to a cell manufacturer working through partners for exactly this reason: it is where the engineering differentiation actually lives.
The investor narrative carries a single logic. The engineering roadmap and the commercial case support each other in every frame. There is no “engineering section” or “commercial section” — the document is one argument, made in two vocabularies.
Why this matters now
Deeptech ventures are being underwritten by investors who have watched the last generation of deeptech companies struggle through the gap between demonstrator and traction. The ones who reach Series A in defensible positions are, almost without exception, the ones that never allowed engineering and commercial decisions to be made on separate tracks. It is the single highest-leverage operating choice a venture can make in its first eighteen months.
What Hatch does
Hatch was built to operate this way. Engineering judgement and commercialisation discipline in the same firm, applied to the same problem, at the same time. Not because it is a differentiator — although it is — but because it is the only operating model that actually works for research-led technology.
Related reading: Why most university spinouts are structurally doomed
