When we started analyzing academic publications for this year’s 𝐒𝐭𝐚𝐭𝐞 𝐨𝐟 𝐃𝐮𝐭𝐜𝐡 𝐓𝐞𝐜𝐡 𝐫𝐞𝐩𝐨𝐫𝐭, one conversion gap stood out.
The Netherlands produces ~40.000 scientific publications each year. Using prediction models, around 7.800 publications show meaningful commercial signals, forming part of the upstream pipeline from which patents, products, and deeptech ventures can eventually emerge.
From that pool, roughly 260 patents are filed annually. And yet, this results in only ~100 new deeptech spin-offs each year.
What makes this dynamic particularly interesting is that spin-off formation looks strong when viewed over a decade. Creation rates have grown from ~62 ventures per year to ~97 annually - a ~1.6× increase, with spin-offs now representing ~9% of startups.
But year-to-year data tells a more nuanced story, with formation rates moderating in the most recent years after peak output in 2022–2023. But even with that acceleration, only a limited share of high-potential science ultimately becomes a company.
Commercialization is also highly concentrated, with most spin-offs emerging from technical universities and health research, where translational infrastructure and venture support are strongest.
And many high-potential discoveries remain at low technology readiness levels, often requiring multiple research breakthroughs, IP consolidation, and founder pathways before venture creation becomes viable.
So the constraint isn’t scientific output. It’s how systematically we identify, mature, and support commercialization opportunities upstream.
The science base is strong. Improving how we convert discoveries into patents, and patents into ventures, is the next frontier for strengthening the deeptech ecosystem.
Read full report: State of Dutch Tech Report 2026
