Why Strong Deep Tech Founders Avoid Accelerator Programmes—and How to Fix the Signal
The founders many deep tech programmes most want are often not in the room.
They looked at the programme, assessed what it had produced and decided that participation would not move their venture forward.
Their absence is usually treated as a recruitment problem. It is better understood as a diagnostic signal about programme design, incentive structures and the evidence available to experienced founders.
What strong deep tech founders look for
Founders with substantive scientific and industrial experience can assess a programme quickly. Partner logos, mentor lists, cohort size and demo-day photographs carry little weight. Follow-on funding totals measure investor activity, not industrial outcome.
What these founders look for is evidence of commercialization:
revenues closed by alumni ventures;
industrial projects that became contracts;
pilots converted into procurement;
licensing deals that improved the founder’s position;
named industrial partners with an operational relationship;
access to decision-makers the founder cannot reach alone.
When that evidence is missing, experienced founders do not necessarily conclude that the programme is bad. They conclude that it is not built for what they need.
Why programmes still fill their cohorts
Seats usually fill because programmes can provide a continuation environment for researchers leaving academia: structured timelines, external funding, peer validation and low immediate pressure between effort and industrial outcome.
That function may be legitimate, but it is different from building ventures capable of industrial conversion. Combining both missions without separating their structures produces a familiar result: ventures that are technically interesting, well presented and optimized for programme participation rather than deployment.
They complete workshops, report milestones and generate visible activity. What they often do not do is cross the industrial boundary the mandate was supposed to address.
The ecosystem race to lower standards
When many programmes compete for a thin pipeline of suitable founders, the competition does not necessarily raise standards. It can lower them.
A programme that maintains evidence-based intake sees cohort numbers fall. One that demands decision-grade technical milestones may see completion rates decline. A programme that relaxes criteria fills seats and stabilizes reporting.
No individual programme needs to choose this outcome. The system rewards activity that survives reporting cycles and punishes rigour that exposes weak fit. Over time, definitions of readiness and progress drift.
Strong founders observe that drift from outside and become more certain that staying outside is the rational choice.
What the empty seats cost institutions
The cost is largely absent from current reporting. Cohorts complete, activities run and funding is distributed. Missing from the dashboard are the founders who decided the programme operated in the capital-and-visibility market rather than the industrial-commercialization market.
The consequence is not one missed applicant. It is a mandate cycle that can continue producing activity without additionality, while technologies with stronger conversion potential build elsewhere or fail to access the right infrastructure.
This pattern connects directly to the evidence in the paper on generic versus specialised support programmes for science and deep tech ventures.
Four changes that improve the signal
1. Report industrial outcomes, not only funding outcomes. Lead with contracts, deployments, procurement and revenue quality. If those outcomes are not yet present, acknowledge the gap and show which structural changes are intended to produce them.
2. Build operational access. A logo is not access. Access means a named procurement, technical or commercial decision-maker who engages because the programme has standing with the organisation.
3. Separate programme missions. Post-academic continuation and industrial conversion require different selection criteria, interventions and success measures. Both can be valuable, but one programme rarely serves both well.
4. Tie renewal to outcome evidence. Oversight bodies should assess whether the programme produced commercialization that would not otherwise have occurred—not simply whether activities were delivered.
How founders should evaluate a programme
Founders can run the same diagnostic in reverse. Review five alumni from recent cohorts and look for customers, deployments, contracts and relevant revenue. Ask which named industrial contacts engage with ventures and on what terms. Calculate the real cost of participation, reporting and equity against the access and evidence the programme has actually produced.
The cost of a poor programme is not only time. It is what the venture becomes during that time: which milestones it learns to pursue, which capital it takes and which industrial relationships it does not build.
This is why strong founders are not anti-programme. They are anti-misalignment.
Programme design should follow the transition
Institutions and founders need the same clarity: a precise reading of what a programme is structurally positioned to produce before either side commits.
I work with funders, programme owners and public institutions to diagnose programme architecture and redesign support around real commercialization transitions. For deep tech founders, the corresponding question is whether a programme protects or reduces the venture’s strategic options.
The absence is the data
When strong founders do not show up, more marketing will not correct the underlying signal. The cohort composition reflects what the programme has demonstrably produced and what experienced founders expect it to produce next.
The system changes when institutions and founders act on the same evidence. It does not change when empty seats are treated only as a recruitment problem.
Kommentare