For Derek Norman, vice president of agriculture venture investments at Leaps by Bayer, the contraction in agtech funding is not just a cyclical downturn. It is a stress test of how the sector develops, validates and allocates capital.
Speaking to AgTechNavigator at the recent World Agri‑Tech Innovation Summit in San Francisco, Norman said the funding environment facing agtech companies today looks markedly different from five years ago.
“Certainly, there’s been a contraction in the number of investors and the amount of capital for companies that are past that seed stage,” he said. “Compared to five years ago, it’s a completely different ecosystem. There’s definitely less money for companies that have done the initial proof of concept or proof of technology.”
With fewer investors supporting a still‑large pipeline of early‑stage companies, Norman believes the sector needs to rethink how it proves value. Central to that rethink is what he calls a fail‑fast mindset.
Testing to fail, not to confirm
At the heart of Norman’s argument is a shift in how agtech companies approach technical validation.
Rather than structuring trials and pilots to show that a technology works, he argues start‑ups should design experiments explicitly to disprove their own hypotheses as quickly as possible.
“The proof of technology work should be approached as: ‘How do I fail this? How do I fail the hypothesis that this technology works as quickly and efficiently as possible?’” he said.
This approach, he argues, forces companies to tackle the hardest questions first – exposing weaknesses early, rather than discovering them later after years of investment.
“If companies do the hardest experiments first and get through them, they’ve moved the ball forward much further in terms of de‑risking,” Norman said. “That lowers the amount of capital they ultimately need to raise.”
For investors, this discipline matters. “From a capital allocation standpoint, that helps investors invest in the right opportunities that have the most rigorous approaches applied,” he said.
Capital efficiency as an existential issue
Norman is clear that this is not simply about better science. It is about the long‑term viability of the agtech investment ecosystem.
“Bringing that mindset to our whole industry will improve capital efficiency,” he said. “And improving capital efficiency is necessary to continue to attract investors into our industry – which is necessary to bring exits, and ultimately to solve the problems we’re trying to address.”
A potential tailwind: the great unbundling of ag
Despite the tough near‑term environment, Norman sees signs that the structural context could improve.
He points to what some have labelled “the great unbundling of ag”. This includes Corteva Agriscience’s plan to split into two independent, publicly traded companies by the second half of 2026; BASF’s preparations for a potential IPO of its Agricultural Solutions division, targeting a standalone ag company by 2027; and Bayer’s own evaluation of its structural options.
“I acknowledge it’s been tough in terms of exits for the past five years,” Norman said. “But this is the first time you can clearly see a set of tailwinds emerging.”
As large incumbents reorganise around growth, he believes there is scope for greater risk tolerance – and potentially more acquisitions.
“These companies may become more growth‑oriented and more willing to use acquisitions as part of that strategy,” he said. That could free up some organisations to be more aggressive, he believes.
Gene editing 2.0: where Norman sees momentum
As head of agriculture venture investments at Leaps, Norman is responsible for identifying high‑impact opportunities – and he says one of the most exciting trends sits at the intersection of gene editing and new computational tools.
“CRISPR tools have really transformed the potential for how you can introduce or improve plants.” he said. “Now there’s a new wave of companies providing additional layers to those tools.”
He cites Pairwise, a Leaps by Bayer portfolio company, as a leader in developing broad, general‑purpose CRISPR toolsets enabling targeted edits across genomes. He also points to Decibel, another Leaps investment, which uses alternative approaches to modulate gene expression without conventional editing.
What’s exciting is the “confluence”, Norman said. “You’ve got CRISPR, you’ve got tools that make editing more efficient, and now you’re bringing in AI and large language models applied to genomes to identify the right changes to make.”
Together, these technologies could accelerate a paradigm shift in how plants are improved.
The Ag Playbook and learning from pharma
Norman is explicit that Leaps by Bayer’s Ag Playbook 2.0 is designed to help enable more of this kind of innovation to make it to market. While he did not write or edit the Playbook, his investment philosophy is clearly embedded in its framework.
The Playbook aims to build “a common understanding of all of the components needed for products to be successful in agriculture,” he said – including manufacturability, regulatory pathways and proof of performance across real farming contexts.
These themes sit behind the sector’s growing debate about whether agtech should borrow lessons from pharma.
“I think there are parts of the comparison that work – and parts that don’t,” Norman said. But he believes pharma gets one thing very right: efficient handoffs.
In pharma, early‑stage companies are routinely acquired once they generate credible early clinical data. Agtech, he argues, lacks a similarly well‑defined transition point. The sector needs to create confidence earlier, he believes – not just that the technology works, but that it can work as a product. “That’s where the playbook comes in… It’s not just proof of technology.”
Corporate capital and the fail‑fast complement
Another live debate in agtech concerns the role of corporate investors in fuelling early‑stage innovation. Asked what advantage Leaps by Bayer offers, Norman emphasised access to operational knowledge.
“You can bring in people from R&D, regulatory or commercial teams who can ask questions start‑ups might not yet be thinking about,” he said – from IP strategy in different regions to manufacturing constraints.
Another ambition of the Ag Playbook, he added, is to standardise that knowledge so it is more widely accessible, rather than locked inside corporate venture arms.
Ultimately, Norman sees the fail‑fast mindset and the Playbook as mutually reinforcing.
“We’ve got a lot of companies and fewer investors,” he said. “Allocating capital to the right opportunities is really important for the viability of the whole sector. Doing the hardest experiments first will help allocate capital to the right opportunities.”
For an industry built on long timelines and high‑stakes science, the message is blunt: learning to fail faster may be the only way to keep moving forward.
