The Pilot Stage: The Difference Between Working and Reliable
Companies at TRL 6 – 7 must shift from proving the possible to proving the repeatable.
The Prototype Stage should end with something that works. The Pilot Stage should end with something that works again and again.
This is the central challenge presented by TRL 6 and 7. A prototype that performed under controlled conditions, tended by the people who built it, in an environment calibrated to show it at its best, is not the same as a technology that performs reliably in conditions for which it did not design, operated by people who did not invent it, in an industrial context that was not arranged for its benefit.
And it is the gap between these two things where a significant proportion of Deep Tech companies stall. From what we have observed and heard about over many years, too many companies confuse the end of the Prototype Stage with proximity to commercial readiness. They are not the same thing.
This essay identifies five stage-features of the Pilot Stage that companies should understand before they enter it.
1. Where the Pilot Stage Sits in the Journey
Before examining the Pilot Stage in detail, it is worth understanding where TRL 6 – 7 sits in the broader risk architecture of the Lab-to-Market journey.
In our essay The Four Risk Phases, we describe how Deep Tech companies navigate four distinct phases, each defined not by a readiness number but by the dominant risk that, if left unresolved, will end the Lab-to-Market journey. The Pilot Stage sits at one of the most important transitions in that framework — the shift from Phase II to Phase III.
Phase II is the phase of Coherence and Translation. Its central question is: can the technology behave predictably, and can the organisation make sense of itself? The science may be plausible, but can it be made to perform consistently, which is to say, reproduced by people other than its inventors, in conditions other than those in which it was originally developed? And can the company bridge the gap between what the technology actually is and what investors, customers, and partners want it to be, without over-committing either side?
Pilot Stage begins in Phase II. Its first task is to resolve that coherence question — to move from ‘it worked in the lab’ to ‘it works when others are in charge.’ But as TRL 7 approaches, Phase III begins to assert itself. Phase III is the phase of Commitment and Constraint. Its questions are harder: can the company make irreversible choices and accept the constraints that come with them?
This is where optionality, which was an asset in Phase I and a reasonable holding position in Phase II, starts to become a liability. A company that is still presenting itself as a platform technology with multiple potential applications at TRL 7 is not in the same position as one that has identified a defensible initial use case and organised itself around it. Credibility in a specific application requires focus and sacrifice, and some of those sacrifices are not easily reversed.
The Pilot Stage, then, is not a single-phase exercise. Companies that navigate it well are those that understand they are completing Phase II and beginning Phase III simultaneously, and that the leadership, commercial, and engineering demands of each are quite different.
The four Lab-to-Market risk phases mapped onto TRL, MRL, and CRL. The Pilot Stage (TRL 6–7) sits within Phase II, with Phase III beginning to assert itself as TRL 7 approaches. Phase transitions are defined by shifts in dominant risk, not by reaching a specific readiness number.
2. What a Pilot Actually Is
There is a version of the pilot that many founding teams carry into TRL 6 that needs to be challenged. In this version, the pilot is essentially a larger, better-resourced prototype, a chance to demonstrate the technology more impressively, to a more senior audience, in a more formal setting. This version of the pilot is not entirely wrong, but it is incomplete.
A pilot is a controlled real-world test of whether the technology can operate at a scale and under conditions that genuinely approach those of eventual commercial deployment. The emphasis here is on ‘genuinely approach’. Not ‘simulate’. Not ‘approximate under idealised assumptions’. The pilot is the first occasion on which the technology encounters something like the actual world — with its variability, its inconvenience, its legacy infrastructure, its human operators, its procurement processes, its organisational sluggishness.
Whereas a prototype is built to demonstrate a concept, a pilot is built to demonstrate operational viability. The question changes from ‘can we make it work?’ to ‘does it keep working when others are in charge?’
This is the Phase II coherence question made real. And companies that grasp it early enough design their pilots accordingly, defining success in terms of reliability metrics and operational consistency rather than peak performance. They make failure modes a central element of the pilot design, and treat the pilot environment as a continued source of learning, not just a demonstration venue.
Crucially, they do not over-promise what the pilot will show. The temptation to frame the pilot as decisive — to position it to investors and partners as the moment that proves commercial readiness — is strong. But this mistake is one of the fastest ways to create a credibility problem.
3. The Pilot Customer
The choice of pilot partner matters.
For many, the chance to pursue a relationship with a large, high-profile organisation that has expressed interest is tempting and occasionally correct. But the qualities that make an industrial organisation impressive — scale, brand recognition, global reach — are not the qualities that make them a good pilot partner, especially for innovation. Large organisations move slowly, have complex approval chains, assign junior contacts to early-stage partnerships, and can change strategic priorities in ways that orphan a pilot with little warning.
From what we have seen and heard from clients, what a first pilot partner actually needs to be is willing, relevant, and able.
‘Willing’ means that there is a genuine internal champion, not someone who attended a presentation and made encouraging noises, but someone who has skin in the challenge the technology is addressing and the seniority to protect the partnership when required.
‘Relevant’ means that the pilot environment is genuinely representative of commercial deployment conditions, not a sandboxed internal experiment that bears little resemblance to the use case the company is actually targeting.
‘Able’ means that the organisation has the technical and operational capacity to run the pilot properly, without requiring the founding team to manage the partner’s internal logistics as well as their own.
The choice of pilot partner is also, in Phase III terms, one of the first genuinely consequential bets the company places on a specific use case and application area. A pilot partner is not just a technical collaborator. They are a signal to the market about where the company has decided to focus. Choosing well accelerates the transition from optionality to commitment. Choosing poorly — or choosing for the wrong reasons — can delay it.
The quality of the pilot customer relationship matters beyond the pilot itself. Industrial partners who have been through a genuine technical engagement — specifically, who have seen the technology struggle, watched the team respond, and observed how the company handles unexpected results — are far more credible advocates than those who simply attended a demonstration. It is these trials that give the entire exercise the credibility required by current and future customers, partners, and investors.
Worth mentioning is the question of data rights, IP boundaries, and what the pilot produces in terms of publishable or shareable results. Such conversations are better had before the pilot begins than after the partnership has started and renegotiation becomes awkward.
4. The Engineering Challenge
The Pilot Stage is where the engineering work changes in character, and where teams configured for research and prototype development can begin to feel the strain.
At TRL 4 – 5, the dominant engineering question is ‘can we make it work?’ The founding technical team — PhDs, post-docs, the PI — are generally well-configured for this. It is, essentially, a harder version of the work they have always done, requiring scientific judgement, experimental ingenuity, and comfort with ambiguity.
At TRL 6 – 7, the dominant question is ‘can we make it work consistently, under variable conditions, in an environment we do not control?’ This is different engineering. It is systems thinking. It is failure mode analysis. It is process documentation. It is the discipline of reproducing results thanks to a process so robust that a competent, external operator can follow it and reproduce it reliably without individual heroics.
In Phase II terms, this is the manufacturing readiness question beginning to assert itself. A company can be at TRL 6 and still be dangerously fragile if its Manufacturing Readiness Level has not kept pace – if the process exists in the hands and instincts of the founding team rather than in documentation and repeatable procedure. This is the binding constraint that Phase II demands be resolved before Phase III can begin in earnest.
From our experience and interviews, many founding teams find this transition genuinely uncomfortable. To many, the work becomes less intellectually stimulating, less creative, far more repetitive. It requires a kind of methodical patience that does not come naturally to scientists accustomed to novelty as their measure of progress.
This is typically the moment when the addition of a Scale-Up Engineering Lead becomes necessary. This profile is distinct from the scientific talent that has carried the technology to this point — someone who has taken technologies from proof-of-concept into operational environments, who understands manufacturing readiness, process control, and the realities of industrial deployment. They are a different kind of person from the founding technical team, and that difference is often the source of productive tension as much as it is a solution.
The founding team’s instinct is to solve engineering problems the way they have always solved problems: through clever thinking. The scale-up engineer’s instinct is to solve them through process design and systematic testing. Both are right. The company’s job is to build a culture in which both instincts are valued, without allowing either to dominate at the wrong time.
5. Leadership at the Pilot Stage
The Pilot Stage can be the moment when the founding CEO’s limitations become meaningfully visible. The risk phase framework helps explain why.
The Phase II leadership imperative is translation: navigating between what the technology actually is and what investors, customers, and partners want it to be, without over-committing either side. The classic Phase II failure mode is allowing external enthusiasm to pull the company forward faster than the engineering coherence supports. This is a genuinely difficult discipline to maintain when industrial partners are expressing interest, investors are pressing for milestones, and the founding team’s own ambition is running ahead of the data.
But as TRL 7 approaches, the need for Phase III leadership starts to assert itself; that is, the ability to decide what the company will not pursue. This is harder than it sounds. Founding teams that have lived through years of possibility — who have maintained optionality as a survival strategy across Phase I and Phase II — can find the act of genuine commitment genuinely painful. Narrowing the field of options feels like loss. But the reality is that maintaining optionality for too long and not making these critical decisions – or worse, hoping partners or clients will make the decision for you – is one of the most reliable ways to stall.
The Prototype Stage also rewards the qualities most founder CEOs carry naturally: vision, scientific credibility, the ability to recruit and inspire, the capacity to hold conversations across technical and commercial domains without necessarily closing them. The Pilot Stage requires the ability to manage execution, set and hold targets, and begin building the internal organisational discipline that scaling a technology demands. It requires the transition from an informal, high-trust, small-team culture to one that is still high-trust but also process-oriented enough to support reliable outcomes.
A good Chair can be particularly valuable here — a stage-appropriate individual with L2M experience who has seen this kind of leadership transition before and who can help the CEO build the commercial and operational muscle demanded, without triggering a defensive and disruptive reaction.
The founding CEO who navigates this well is one who has genuinely accepted that their role is changing. The company now has more people, more moving parts, and more external stakeholders who need to be managed with consistency. The CEO who remains the company’s best scientist is not the same CEO who builds the company’s first reliable process. The best Pilot Stage leaders find a way to be both, or to build around themselves with enough honesty and speed that the gap never becomes a bottleneck.
6. Capital and What Investors Are Actually Buying
Pre-seed and seed investors, broadly, are buying the founding team, the technology concept, and a credible hypothesis about where the technology fits. They are investing in the possibility of the journey.
The Series A investor is buying evidence of operational viability — evidence that the technology holds up outside the lab, that the team can execute under real conditions, and that the commercial pathway has been tested against industrial reality. In Phase II language, they are assessing whether the company has answered the coherence question: can the technology behave predictably, and has the organisation begun to make sense of itself?
This is why the quality of the pilot matters so much to the capital conversation. A pilot designed to impress — ‘ideal’ conditions, peak performance, carefully managed access — produces a story. A pilot designed to generate operational evidence — including the challenges encountered, the failure modes discovered, and the design iterations that followed — produces credibility. Serious Deep Tech Series A investors know the difference.
The rest of the capital stack at this stage typically includes grant funding and, where the pilot partner relationship has deepened appropriately, early forms of commercial revenue. An organisation that has moved from pilot partner to early paying customer, even if it is not repeatable revenue, is a powerful signal to carry into a Series A conversation. It is evidence that someone other than an investor has examined the technology in an operational context and decided it is worth paying for.
Government and public funding also continues to be available at this stage in many geographies, particularly for technologies with national strategic relevance. Such funding is not always the easiest to access, and can come with reporting obligations that add to the management burden. But it is worth pursuing as a validation signal that carries weight with private investors performing their own diligence.
Companies that arrive at the end of the Pilot Stage with a completed pilot, a real path to a commercial-grade F.O.A.K., a documented set of learnings, an industrial partner relationship that has survived genuine operational challenge, and a Series A investor brought along through that process — these are the companies that have genuinely earned the transition to TRL 8.
Any reward, however, can feel short-lived. Because what comes next — the First-of-a-Kind Stage — is where Phase III demands its full due. The stakes increase again, and where the lessons either learned or not learned, and the behaviours either adopted or not, become the difference between a company that scales and one that discovers it was not quite ready.
Deep Tech Leaders is building the ‘data-first’ operating manual and talent network for companies navigating the Lab-to-Market journey. Through domain-specific data & insight, long-form analysis, in-depth conversations, and our world-leading talent network, we surface how Deep Tech companies actually progress – and why so many fail. This work is underpinned by our executive search practice focused on placing proven leadership talent into roles where phase, risk, and capability align.
