Biotech startups get funded in stages, and each stage buys a specific kind of proof. Seed money pays for discovery work and early lab data. Series A pays for preclinical development and the studies that support an investigational new drug application. The hardest stretch sits between them: a company with promising science but no human data often cannot raise the next round, a gap investors call the valley of death.
The reason is structural. A software company can ship a product in months and let revenue justify the next raise. A drug or device company needs years of experiments before anything sells, so each financing round has to be justified by a milestone rather than by sales. Investors price those milestones, and founders plan their spending around them.
Biotechnology itself is a broad discipline. As Wikipedia's overview of biotechnology describes it, the field integrates natural sciences and engineering sciences to apply organisms, or parts of them, to products and services, ranging from medicines to biofuels and genetically modified crops. That breadth matters for funding, because a company engineering microbes for industrial output follows a different capital path than one developing a cancer therapy. This piece focuses on the therapeutic and diagnostics path, where the staged model is most rigid.
What does seed funding actually buy in biotech?
Seed capital buys the first credible evidence that a scientific idea works. That usually means validating a target, showing a molecule or engineering platform does something measurable in cells or animal models, and assembling the core team. The deliverable is not a product. It is a data package good enough to convince a specialist investor that the next stage is worth pricing.
Seed checks come from a recognizably different set of pockets than later rounds. Common sources include:
- Angel investors with backgrounds in life sciences, who can judge early data themselves.
- Specialist seed funds that focus on biotech and accept the long timelines.
- Government grants, such as small-business research awards, which provide non-dilutive money that does not cost the founders equity.
- University licensing and spinout structures, where the institution contributes intellectual property and sometimes early capital.
The amount raised at seed varies widely by modality and geography, and no single figure describes the market. What stays constant is the expectation: seed money is spent on generating the preclinical story, not on building a company infrastructure the science cannot yet support.
What does a Series A require that a seed round does not?
A Series A investor wants a defined development plan, not just a hypothesis. By this stage the company is expected to show reproducible preclinical evidence, a clear regulatory strategy, and a named indication. The round funds the work that turns lab science into a regulated development program: formulation, toxicology planning, manufacturing development, and the design of first-in-human studies.
The investor base shifts accordingly. Series A rounds are typically led by venture firms that specialize in life sciences and employ people who have run drug development or managed regulatory filings. These firms negotiate for board seats and milestone-based tranches, meaning capital is released as the company hits agreed checkpoints rather than all at once. Readers following this should also see Tandem Health Raises $100M Series B Led by EQT-Managed Scaleup Europe Fund.
That structure cuts both ways. Tranches protect investors from a failed experiment, but they also give the board leverage to redirect or wind down a program. Founders coming out of academia often underestimate how much the Series A is a governance event, not just a bank transfer.
Why is the gap between proof and IND called the valley of death?
The valley of death in biotech is the period after a company has convincing discovery data but before it has the safety and manufacturing package regulators require before human trials. The work in that stretch, often called IND-enabling studies, is expensive, unglamorous, and produces no headline results. Yet without it, no regulator will allow a first-in-human trial, and without the prospect of human trials, later-stage investors lose interest.
Several forces converge in this gap. Preclinical programs cost more than seed budgets can cover. Public markets and crossover funds generally wait for clinical data. Grant funding rarely stretches to full toxicology packages. And corporate partners, which can be a lifeline, usually want to see at least some clinical signal before committing. A company that cannot bridge its own funding gap has limited options: shrink the program, license the asset early on weak terms, or shut down with the science still untested in people.
The gap is not unique to drugs. Diagnostics and medical device startups face a version of it between prototype validation and regulatory clearance, where the engineering is done but the clinical evidence and manufacturing scale-up remain unfunded. Readers tracking how device and diagnostics companies navigate that stretch can see the pattern in current coverage of point-of-care molecular diagnostics launches, where the leap from validated assay to deployed product is the defining challenge. We covered a connected angle in What Recent Point-of-Care Molecular Diagnostics Launches Mean for Decentralized Testing.
How do companies bridge the gap?
Founders use a mix of tools, and the strongest companies stack several rather than relying on one.
- Non-dilutive capital. Grants, contract research partnerships, and government programs fund specific work without selling equity. They rarely cover a full IND package, but they extend runway.
- Early licensing or co-development. A larger company may pay for option rights on an asset. The terms are usually favorable to the partner, because the risk is highest at this stage, but the money keeps the program alive.
- Venture creation models. Some funds build companies in-house, staff them, and fund them through the preclinical period on a portfolio basis, accepting that most will fail.
- Staged milestones with existing investors. A syndicate that believes in the platform may fund the IND-enabling work directly, priced against the regulatory filing rather than a clinical readout.
What this means for founders is that the choice of first investors shapes the options available in the valley. A seed syndicate with deep life-science pockets can follow on. A generalist syndicate often cannot, and the company discovers that late.
What signals do investors read before writing a check?
Across stages, specialist investors look for the same few things, weighted differently by round.
- Data quality. Reproducible results, appropriate models, and honest handling of negative findings carry more weight than any single dramatic experiment.
- Intellectual property position. Patents and exclusivity determine whether a future product can earn enough to repay a decade of development.
- Team credibility. People who have taken a product through regulation before reduce execution risk, which is the dominant risk at Series A.
- Capital plan realism. A budget that assumes the next round arrives on schedule, with no contingency, reads as a red flag.
Platform companies, such as those connecting computational design with experimental validation, add a further consideration: whether the platform generates repeatable value across targets or depends on one lucky asset. Coverage of AI molecular design paired with experimental validation illustrates how investors evaluate that question in practice.
What the financing path establishes, and what it does not
The staged model exists because biology resists shortcuts. Seed proves the idea can work in a lab. Series A proves a team can turn lab work into a regulated program. Clinical stages, which sit beyond this piece, prove the product works in people. Each gate filters, and each filter is expensive.
What the model does not solve is the timing problem at its center. The valley of death persists because the most valuable work, the IND-enabling package, is the least fundable work in the ecosystem. Until that changes, the companies that survive will be the ones that plan for the gap from day one, stacking non-dilutive money, partnerships, and patient capital before they need them.
For ongoing coverage of how funding, platforms, and policy interact across the sector, the publication's tech news section tracks the deals and the misses as they land.
