One question separates a biotech investment from a gamble — and most investors ask it too late: what stage is the science actually in?
Not what stage the pitch deck says. What stage the data proves.
This guide breaks down the four investment metrics that institutional biotech funds actually use, the three red flags that kill deals at Series A, and the public data sources that let any investor verify a founder's claims before writing a check.
Biotech investing is a probability game. The global biopharmaceutical market is worth over $700 billion, growing at 8% CAGR through 2030. A single FDA approval can generate billions in revenue. But only about 10-12% of drugs that enter Phase 1 clinical trials ever reach the market. The other 88-90% fail — and with them, most of the capital invested.
The asymmetry is brutal. A Phase 3 readout that hits its endpoint can 10x a company's valuation overnight. A miss can wipe out 80% of value in a single trading session. The difference between those outcomes is not luck. It is whether the investor knew which data points actually de-risk a therapeutic asset — and which ones are just decoration.
The Four Metrics That Actually De-Risk a Biotech Investment
Institutional biotech investors — ARCH Venture Partners, OrbiMed, Third Rock Ventures, Versant Ventures — evaluate startups on four dimensions. Every other question is secondary.
1. Clinical Stage
Clinical stage is the dominant de-risking milestone in any biotech. The progression is linear: IND filed → Phase 1 → Phase 2 → Phase 3 → FDA submission. Each transition cuts the risk of total capital loss by roughly half.
A preclinical company with no human data is a fundamentally different investment than a company awaiting Phase 3 readout on a drug that has already shown efficacy in 200+ patients. The success rate from Phase 1 to approval is ~10-12%. From Phase 3 to approval, it jumps to ~58%. Investors price that difference into valuations — but not always accurately.
The question to ask: What is the specific endpoint at the next stage, and what data do you already have that predicts it? If the answer is "we will know after the trial," the risk is still fully priced in.
2. Target Patient Population
Broad indications (diabetes, hypertension, major depressive disorder) address millions of patients but face the heaviest regulatory scrutiny, the longest trials, and the most crowded competitive fields. Rare disease indications (fewer than 200,000 US patients) qualify for FDA orphan drug designation — faster review, tax credits, and seven years of market exclusivity. The trade-off is a smaller commercial ceiling, offset by higher pricing power and lower trial costs.
The best risk-adjusted returns in biotech over the last decade have come from companies that target well-defined orphan populations with a clear biomarker and an accelerated regulatory path. The worst returns have come from me-too drugs chasing million-patient markets with no differentiation.
3. Preclinical Efficacy Data
In vivo (animal) studies in validated disease models are the floor for credible science. In vitro data alone — cells in a dish — tells you almost nothing about whether a therapy works in a living system. But not all animal models are equal. Rodent data has famously poor predictive value for human outcomes. Investors have been burned repeatedly by companies with pristine mouse studies that evaporated in Phase 2.
The signal to look for: efficacy data in large-animal models (non-human primates, pigs) or in humanized mouse models that carry transplanted human tissue or immune cells. These are significantly harder to game and correlate better with clinical outcomes.
4. IP Protection
Composition-of-matter patents — patents on the molecule or therapy itself — are the foundation of biotech IP strategy. They protect the asset, not just the method of making it. Method patents are weak. If a startup only has method patents and no composition-of-matter protection, the molecule can be legally copied by any competitor that finds a different way to produce it.
Patent expiry timeline determines the commercial window. A drug with 12 years of exclusivity is worth roughly double one with 6 years, all else equal. The clock starts ticking from the earliest filing date — not from FDA approval. Every year spent in clinical trials is a year of patent life consumed before a single dollar of revenue is generated.
Three Red Flags That Kill Biotech Deals
Experienced biotech investors scan for these three patterns before they even read the financial model. Any one of them is grounds for a pass.
🔴 No Composition-of-Matter Patent
If a biotech startup's only intellectual property is method-of-use or process patents, the asset has no durable moat. A large pharma company with a better manufacturing process can legally develop the same molecule and beat the startup to market. This is the single most common reason institutional biotech funds walk away after initial screening.
🔴 Animal Data in Non-Validated Models
Using a proprietary mouse model that the startup itself developed and that no independent lab has replicated is not evidence. It is a circular reference. Credible preclinical data comes from established, published disease models that the scientific community accepts as predictive. If the model does not appear in any peer-reviewed literature outside the startup's own publications, the data should be discounted to near zero.
🔴 Key Scientist Departs
Biotech investing is inseparable from the scientific founder. When the principal investigator who conceived the therapeutic approach leaves — or is not committed full-time — the clinical credibility of the entire pipeline collapses. Unlike SaaS, where a good CEO can hire a new CTO, biotech intellectual capital is personal. The science lives in the scientist's head. If they leave, the company's core asset is a patent that the new team may not know how to prosecute.
Funding Round Expectations by Stage
Biotech venture capital operates on a different cadence than enterprise software. A typical biotech startup will raise 3-5 rounds before reaching revenue, with each round tied to a specific clinical milestone rather than a growth metric.
Pre-seed and seed rounds ($2-15 million) fund target identification, assay development, and initial animal studies. The company at this stage has no human data and no regulatory filings. Valuation is driven by the founding team's scientific reputation and the size of the addressable patient population. Serial biotech founders with prior exits can command 2-3x the valuation of first-time founders at the same scientific stage.
Series A rounds ($15-60 million) typically fund IND-enabling studies — the toxicology, pharmacokinetics, and manufacturing data required to file an Investigational New Drug application with the FDA. A well-prepared Series A biotech has already had a pre-IND meeting with the agency and incorporated its feedback into the trial design. Companies that skip this step seldom get a term sheet from top-tier biotech funds.
Series B ($40-100 million) funds Phase 1 and Phase 2 clinical trials. This is where most of the value destruction happens. Phase 1 safety data that looks clean can double a company's valuation. Phase 2 efficacy data that misses its primary endpoint can destroy it entirely. The biotech funding market confirms this pattern: over the 12 months ending June 2026, late-stage rounds (Series B and beyond) accounted for 60% of the $16.976 billion raised, reflecting investor preference for de-risked clinical assets over preclinical speculation.
Series C and later rounds ($80 million+) fund Phase 3 registrational trials and commercial manufacturing scale-up. At this stage, the risk is no longer scientific — it is commercial. Will the drug get broad label approval? Will payers reimburse at the projected price? Will a competitor's parallel Phase 3 readout reshape the standard of care before launch?
Mapping the Competitive Landscape
A biotech startup's competitive position is defined by three factors: mechanism of action, clinical stage, and regulatory pathway. Two drugs targeting the same disease through different mechanisms are not direct competitors — they may even be complementary. Two drugs using the same mechanism at the same clinical stage, however, are in a winner-take-most race.
The comparison frame for a due diligence investor: the market expects a biotech to either be first to a validated mechanism, or to demonstrate clearly superior efficacy/safety data in a head-to-head trial. "We are a fast follower" works in enterprise software. In biotech, it means you arrive after the standard of care has already been set — and your drug now needs to beat an entrenched therapy with years of real-world data behind it.
Regulatory & Compliance Risks
The FDA pathway is the binding constraint on biotech value creation. Every clinical phase is a regulatory gate: the agency can halt a trial, demand additional data, delay a review, or convene an advisory committee that recommends against approval. These events are binary catalysts — they either unlock or destroy the next tranche of value.
The documents an investor should verify before committing capital: pre-IND meeting minutes with the FDA (showing the agency's feedback on trial design), IRB approvals for ongoing trials, and any FDA warning letters or clinical holds in the company's history. An FDA hold — even one that was later resolved — is a permanent stain on a company's regulatory track record.
Signals Worth Tracking
Four public data sources let any investor verify a biotech startup's claims without relying on the pitch deck. Institutional due diligence teams check all of them as standard procedure.
ClinicalTrials.gov — registered study status and enrollment progress. A trial that is "not yet recruiting" six months after the listed start date is a red flag.
PubMed — peer-reviewed publications from the founding team. If the science is real, it has been published.
USPTO patent database — core composition-of-matter patents. Verify filing dates, inventor names, and patent family scope.
FDA warning letters database — any prior FDA actions against the company, its CROs, or its manufacturing partners.
The biotechnology market raised $16.976 billion across 128 disclosed deals between July 2025 and June 2026, with a median round size of $100 million. Therapeutics discovery platforms accounted for 77.6% of all disclosed capital. North American companies captured 63.5%. The data confirms the framework's signal: capital is flowing to assets with clear clinical stage progression, not to preclinical speculation.
The investor who evaluates a biotech startup by asking the right questions — about the science, the patent, the regulatory pathway — before looking at the valuation is the investor who survives the 88% failure rate that defines this sector.