Science Policy · Guest article · jmwiarda.de, 7 July 2026 · https://www.jmwiarda.de/blog/2026/07/07/wenn-exzellenz-steckenbleibt
Between a discovery and a product, what is missing is rarely the idea and almost never the money. Yet too much good science seeps away in competing responsibilities, project logics and a lack of operational continuity. A guest article by Michael J. Bojdys.
A discovery that works in the lab but reaches no one is not half a success. It is sunk taxpayer money with a publication as an appendage. I have spent nine years in Berlin between a research group, a university alliance, several consortia, a spin-off and the committees in which transfer rules are written – and the same pattern showed at every station. What keeps good science from becoming a product is rarely the idea and almost never the budget. It is the layer in between: who decides, on what evidence, by when, and who carries the consequences when a deadline slips? The path from idea to product is a chain of handovers. Every handover needs someone who answers for it, and every responsibility needs a structure that survives the next change in personnel. Transfer breaks at whichever link is missing first.
On 1 August I leave Berlin for Brussels to fund high-risk ventures at the European Innovation Council – the EU’s funder of high-risk technologies. Before I go, I want to pass on to those who stay and keep building Berlin as a science and technology location, as an innovation ecosystem, what took me a long time to recognise myself.
Where time actually goes
Ask why a chemical result takes ten to fifteen years to reach the market and you will hear about technical risk, scaling and regulation. These factors are real. But most of the time is not lost there. It is lost because knowledge stays locked in individual labs, formats and jurisdictions, without shared infrastructure that makes a result findable, verifiable and usable by the next group. An insight that one team cannot exploit on its own is left lying, because there is no path for handing it to a team that could.
This gap is not just a problem of individual labs. It also shows at the level of entire locations: wherever big goals are proclaimed but operational continuity is neither co-funded nor bindingly organised. What fails at the handover in a single lab fails at location level at the other end of the chain: at continuity. Berlin is an instructive case. The Berlin University Alliance kept its national excellence status in 2026 while roughly of once-pledged core funding were missing; the reserves of the three universities are likely to be exhausted by 2028, and the main building of the Technical University has been closed since 9 May 2026. These are not simply three crisis headlines side by side. They show what happens when excellence, infrastructure and durable capacity to act come apart. Excellence is pledged, day-to-day operations are unfunded, and responsibility for this gap has no address.
A consortium fails the same way as soon as its shared decision history disappears and every email turns into an audit of memory and motives. Transfer fails quietly, one stalled handover after another, and no one is held responsible.
“A discovery that reaches no one is not even half a success. It is sunk taxpayer money with a publication as an appendage.”
The asymmetry no one mentions
Continuity in the German science system is not a question of need but of legal form. And even though individual research and transfer projects at universities almost never fail for lack of budget, the funding logic is still part of the problem. Transfer needs permanent functions, a framing that sits well in no project logic: people who document decisions, settle IP questions, keep data interoperable, know the regulatory pathways, and do not start again from zero after a change in personnel.
The four large non-university research organisations can provide that. Max Planck, Helmholtz, Fraunhofer and Leibniz, together with the DFG, have a guaranteed annual uplift of three per cent, fixed until 2030 in the Pact for Research and Innovation. Universities, where most early discoveries and nearly all doctoral training take place, have no comparable guarantee for their core funding; the federal funds for their teaching, the Zukunftsvertrag, are smaller, rise only moderately, and how things continue after 2027 is still to be negotiated between the federal government and the Länder. As a result, transfer capacity migrates to the institutions with the more stable base, which helps explain why so much science originating at universities finds no way into application. The German Rectors’ Conference has demanded this parity for years, without success. The guaranteed funds are thus tied to the structures, not to the mission – and that is exactly where the gap opens.
A shared yardstick
If the operational layer is not to remain invisible, it needs a shared yardstick, because responsibility can only be assigned once it is settled in advance what counts as evidence. A proven yardstick exists. In almost every conversation about a project or a consortium I ask the eight Heilmeier questions, named after former DARPA director George H. Heilmeier:
- What are you trying to do? Describe the objective in everyday language, without jargon.
- How is it done today, and where are the limits of current practice?
- What is new in your approach, and why will it succeed?
- Who cares? If it succeeds, what difference does it make, and for whom?
- What are the biggest risks – technical, regulatory, market, financial, within the team?
- What will it cost? A rough order of magnitude for the next three to five years.
- How long will it take – to first impact and to full success?
- How will you measure success? Concrete interim and final checks.
The questions sound simple and are very memorable, and that is precisely their strength in a stakeholder conversation: the architects of an innovation ecosystem must measure themselves and their partners against them, and do so again at every progress report. A team that cannot answer question 4 has a publication, but no project. A consortium that dodges question 8 has a press release, but no plan.
What already works
Germany and Europe do not lack instruments. They lack the connecting practice between them. The building blocks of a solution are in place.
Germany has created transfer formats such as T!Raum and the DATIpilot; SPRIND funds leap innovations, the European Innovation Council funds high-risk ventures across Europe, and FAIR infrastructures are meant to make research data findable and interoperable. On the digital substructure, the National Research Data Infrastructure (NFDI) has come furthest, funded with up to 90 million euros a year from the federal government and the Länder until 2028. What has emerged, however, are more than two dozen consortia, neatly separated by discipline, very uneven in quality, and the leap into industrial application has so far largely failed to materialise. Many data islands, few bridges between them. It sets standards and secures data; it does not organise the concrete handover in the individual project.
The German Science and Humanities Council reaches the same finding in its evaluation of July 2025. Project-shaped funding, it holds, carries no permanent infrastructure and does not retain the specialist staff; the governance bodies are too complex to steer. It recommends consolidation, permanent positions and a rebuild of the governance, and funding from 2029 onwards is yet to be decided by policymakers. So the largest shared data project stands where the universities stand: needed permanently, funded on fixed terms.
None of these elements replaces the operational translation between a result and the team that could make something of it. That takes named people, decision rights and the habit of being present where funding logics, standards and transfer rules are made. In 2019 – then in the circle of the World Economic Forum’s Young Scientists – I brought proposals for accelerating transfer to the members of parliament driving SPRIND and DATI in the Bundestag, from government and opposition alike. Ahead of me in the corridor a lobbyist from the pharmaceutical industry was waiting, behind me another lobbyist from the chemical industry, and apart from me no scientist stood in the line. The instruments were built. The habit of showing up where they are designed did not follow.
Four steps for those who stay
None of these steps requires a technical breakthrough. What they do require are innovation-ecosystem architects who take responsibility, beyond the first setback. My conclusions:
- Build the operational layer before you need it. Fix who has decision authority, what counts as evidence, where the thresholds for “go” or “stop” lie, and how fast you act in a crisis. The eight questions above are a good start for what counts as evidence. Then begin with one page for a project: the one person who can stop it, the two or three findings that justify a stop or a change of course, and the place where that decision is recorded, so it is not reopened by email a month later. This agreement is the smallest working version of the whole operational layer, and most teams have never made it.
- Keep knowledge in common. Put research results on a shared, FAIR-compliant infrastructure, so that a result is findable and usable beyond the lab as well, and a stalling finding can migrate to a group that pushes it forward. A patent lying unread in a single institute produces no transfer.
- Pool competences before you are forced to. Berlin’s reflex is to send every chair, every institute and every university to the ministry or the Senate on its own. Start with a one-line mission the partners genuinely agree on, and with the one competence the pooled alliance offers that none can deliver alone. If you cannot name that competence, you have an administrative merger that will not survive the first budget dispute.
- Be where the rules are written. Transfer policy, funding criteria and standards arise in rooms that scientists rarely enter. Put one knowledgeable person into each of them – a committee, a hearing or a standards body – and decisions will no longer be made only by those who sell into them.
What it costs to ignore this
Treating the operational layer as overhead is the expensive choice, and the bill arrives late. A discovery that has to wait does not announce its own death; it simply never becomes the company, the therapy or the material it could have been, while the funds that paid for it quietly turn into a citation metric. Without continuity, no one takes responsibility for long. Without responsibility, no handover happens, and the idea stays stuck in the nursery of the universities. Those who stay in Berlin decide where this goes. The operational layer that carries science into the world must be standing before anyone is forced to it – because life, as the saying goes, punishes those who come too late.
About the author
Michael J. Bojdys is a chemist, inventor and innovation adviser; until the end of July 2026 he leads the Functional Nanomaterials group at Humboldt-Universität zu Berlin, where he co-founded the battery spin-off “MANA.energy”.
Sources
- Funding gap and building closures at Berlin universities: Written inquiry by T. Schulze (The Left Party), November 2025, via Jan-Martin Wiarda’s blog.
- Pact for Research and Innovation IV (3% annual growth, 2021–2030), BMFTR.
- Future Contract: Strengthening Studies and Teaching, BMFTR.
- German Rectors’ Conference on the Higher Education Pact and Future Contract.
- The Heilmeier Questions (Heilmeier Catechism), DARPA.
- FAIR data principles, GO FAIR.
- SPRIND, Federal Agency for Disruptive Innovation.
- RM Morgan et al., The missing link of science in policy, Frontiers Policy Labs, 2022.
Innovative research has brought us closer to sustainable battery technology with a breakthrough in sulfur-based cathodes. Traditionally, lithium-ion batteries—central to electronics and electric vehicles—rely on scarce materials like cobalt. Sulfur offers a greener alternative due to its abundance and impressive theoretical capacity of 1675 mAh g-1.