Leadership for deep tech: why standard search playbooks fail

Hiring leaders for deep tech, companies whose progress depends on engineering and physical science as much as software, is not the same as hiring for a consumer or SaaS business, and the standard search playbook quietly fails at it. The failure is rarely dramatic. It is a slow drift toward plausible candidates who do not actually fit, because the methods that work for software hiring do not reach the people deep tech needs.

The talent is small, dispersed and non-obvious

Deep tech talent pools are tiny and globally scattered. The leader who can take an advanced material from lab to plant, or commercialise a hard physical technology, does not sit in a large, easily searchable pool. They are often inside large industrial companies, national labs, or a handful of specialised startups, and they are rarely looking. A search that relies on the usual channels, the obvious startup shortlist and the open market, simply will not find them.

Worse, the right profile is frequently non-obvious. The best commercial leader for a deep tech company might be a sustainability leader from an industrial conglomerate or a business-development leader from an energy firm, not a polished startup executive. Finding them means searching adjacent fields deliberately, which a generic process does not do because it does not know to look there. The candidate who will actually succeed often does not match the résumé the company first imagines.

Credibility runs in two directions

Deep tech leaders usually need to be credible to two audiences at once: the scientists and engineers inside the company, and the industrial customers, partners or regulators outside it. A leader who commands one but not the other struggles. The commercial leader who cannot earn the respect of the technical team cannot lead it. The technically credible leader who cannot sell to a sceptical industrial buyer cannot grow the business. This dual credibility is rare, and it is hard to assess through a conventional interview that probes only one side.

Time horizons compound the difficulty. Deep tech builds over years, under constraints of capital, regulation and physical reality that software does not face. Leaders have to hold a direction through long cycles without losing momentum, which is a different temperament from the fast-iteration instinct that consumer tech rewards. Confusing the two produces leaders who are impatient with the very timelines their company depends on, and that impatience can do real damage to a company whose value is built slowly.

What the work actually requires

Search for deep tech leadership has to start from the science and the market, not from a job title. It means mapping global clean tech, industrial, materials, energy and frontier ecosystems, depending on the domain, and understanding who is genuinely doing the work. It means defining the role around the specific technical and commercial reality of the company, because deep tech roles do not generalise. And it means assessing for dual credibility, risk appetite, and comfort with long horizons, none of which appear cleanly on a resume.

Assessing for the long horizon

The hardest thing to assess in a deep tech leader is temperament for the long haul, and it is also the thing that matters most. A leader who is brilliant but quietly impatient will erode a company whose value is built over years, pushing for visible wins that pull the science off course or burning out when the timelines refuse to compress. Conventional interviews rarely surface this, because every candidate says they understand long horizons. The signal comes from history and from references: have they actually stayed with hard, slow problems before, or does their record show a pattern of leaving once the early excitement fades? How did they behave in the difficult middle of a long build, when progress was real but invisible to outsiders?

Dual credibility needs the same evidentiary care. Rather than asking whether a candidate can earn the respect of both scientists and industrial buyers, the better approach is to test it: have the technical team probe them, and check with the kind of customer or partner they would need to win. The leaders who pass both tests are rare, and finding out before the offer, rather than after, is most of the value a serious search adds. None of this is fast, but deep tech does not reward fast hiring. It rewards the right hire for a long road.

We have placed business and technical leaders into deep tech companies precisely by working this way, targeting operators who had crossed from large corporates into early-stage roles, and searching non-obvious pools rather than the obvious shortlist. The standard playbook fails at deep tech not because the people do not exist, but because the method was built for a different kind of hiring. Deep tech needs a search built for its own reality, and that is the work.