Deep-science financing does not follow the software cadence. CellCentric's record — $220M raised, concentrated in a single Series D backed by specialist life-science funds, a pharma venture arm and mission capital — reflects binary milestones: you fund the whole readout or you do not start it, and you raise on the far side of the event with runway to spare.
Key takeaways
- Science risk retires in jumps, so rounds are lumpy rather than laddered.
- Fund past the binary milestone, not up to it, or you negotiate from zero leverage.
- Specialist funds, pharma venture arms and mission capital each diligence on different criteria and timelines.
- Milestone tranching is standard in these categories and is not a signal of weak conviction.
- Any founder facing a binary milestone — approval, certification, one anchor contract — should use this cadence, software or not.
Most fundraising advice online is written for software: raise 18 months, show growth, raise again. Apply that to a company whose next milestone is a clinical readout or a working plant and the plan breaks immediately.
Rather than generalising, this works through the differences using a documented record: Tomasz Knurowski at CellCentric (Little Chesterford, United Kingdom), which develops targeted epigenetic therapies.
| | | |---|---| | Company | CellCentric (Little Chesterford, United Kingdom) | | Total raised | $220M | | Latest round | Series D — $220M | | Round date | July 2026 | | Named backers on record | RA Capital Management, Forbion, Morningside, Pfizer Ventures, Avego, American Cancer Society |
The striking feature is the shape: the disclosed lifetime total and the latest round are the same figure in this record — a concentration profile, not a ladder. That is characteristic of categories where value steps up at discrete, expensive events.
Software risk retires continuously — every month of usage data reduces uncertainty a little. Science risk retires in jumps. A readout either happens or it does not; a plant either commissions or it does not. There is no partial credit.
The financing consequence: you cannot raise a small round to make partial progress toward a binary event. You raise the full cost of reaching the event, or you do not start it.
Three investor types appear in records like this one and rarely in software rounds:
Specialist life-science funds with the scientific staff to underwrite the mechanism itself.
Corporate venture arms of pharmaceutical companies, which are simultaneously potential partners and potential acquirers.
Mission-aligned and non-profit capital, which underwrites the disease area rather than the return profile alone.
Each diligences differently. Specialists will interrogate the data package for months; corporate arms move at internal committee speed; mission capital cares about the indication and the trial design.
Start 12 months out, not six. Scientific diligence is slower than commercial diligence, and it involves external experts you do not control.
Fund to the readout, plus the time to raise on it. A round that ends the month results land leaves you negotiating from zero leverage.
Expect tranches. Milestone-based tranching is normal here and is not a sign of weak conviction; the second tranche is priced off risk you have not yet retired.
Handle the strategic carefully. A pharma venture arm on the cap table is valuable and informative. Its diligence file is also a competitive read on your programme. Scope what it sees.
Even if you build software, ask whether your next milestone is continuous or binary. If it is binary — a regulatory approval, a single anchor contract, a hardware certification — then the software financing cadence is wrong for you and this shape is the better template.
Amounts, stage, date and named participants are documented. Tranche structure, valuation and terms are not. Read shape, not price.
Frequently asked questions
- Why are biotech rounds so much larger than software rounds?
- Because value steps up at discrete, expensive events. There is no partial credit for getting halfway to a readout, so the round has to cover the whole event.
- How early should I start a deep-science raise?
- Around twelve months out. Scientific diligence involves external experts and committee timelines you do not control.
- Is a tranched round a bad sign?
- Not in these categories. Tranches price the risk that has not been retired yet and are a normal structure rather than a mark against the company.
- Should I let a pharma venture arm diligence my programme?
- Often yes, with scope limits. The capital and validation are real, but the diligence file is also a competitive read on your work.
- Where do the figures in this article come from?
- From the structured founder funding records we maintain: total raised, round stage, round amount, round date, and named participants. They exclude valuation, deal terms and board composition.
