Pharma partnership terms that protect your pipeline when the science goes sideways
Milestone-based deals are how most biotech companies survive long enough to see their drug approved, but poorly structured agreements can leave a CFO holding the downside while the partner captures the upside. This playbook shows how to build deal terms that align incentives across a decade-long development arc.
Turing LedgerFinance & Strategy AnalystSeptember 27, 2026Listen to the podcast
4 min
Chapters
Key takeaways
- Risk-adjust every milestone using real stage by stage transition probabilities, starting from the 5 to 7 percent Phase 1 to approval rate in oncology.
- Insist that reversion rights return the clinical data and safety file, not only the compound and patent.
- Write diligence obligations with minimum annual spend and fixed development dates, with the licence reverting if the partner misses them.
- Spend your leverage at signing, before the ink dries, since that is the only moment you hold the asset outright.
- Trade away upfront cash if you must, but never trade reversion, data return or diligence obligations.
Read the full transcript
Host:Leaders Insights. Today: Pharma partnership terms that protect your pipeline when the science goes sideways. Three weeks ago a mid-cap biotech watched a partner walk away from a Phase 2 program and kept exactly nothing — no data rights, no reversion, no cash. That's the nightmare, right?
Expert:It's the nightmare, and it's avoidable. The deal was signed in 2021 with a clause that let the partner terminate "for convenience" with ninety days' notice and no obligation to hand back the clinical data. So the biotech funded four years of development and ended up with a hole where their lead asset used to be.
Host:But every founder tells me milestone deals are how they survive. Cash upfront, cash at each stage. What's wrong with that?
Expert:Nothing's wrong with the structure. The problem is the biotech reads the headline number and ignores the shape of the payments. Someone signs a deal worth, say, 900 million dollars in "biobucks" — total potential value across every milestone — and 40 million of that is actually upfront. The other 860 is contingent on approvals that hit maybe one time in ten.
Host:So the nine-figure press release is mostly fiction.
Expert:Mostly aspiration. And here's the number that matters: in oncology, roughly 5 to 7 percent of drugs entering Phase 1 — the first human safety trials — ever reach approval. So if 90 percent of your deal value sits behind a Phase 3 readout, you should price it at pennies on the dollar in your own model, not at face value.
Host:Give me the term that actually protects the downside, then. Not theory — a clause.
Expert:Reversion rights. If the partner shelves or terminates the program, the asset and all the data come back to you cleanly, with a defined licence to anything they generated. Vertex structured deals like this years ago — walk away, and the science reverts, no ransom. That single clause is the difference between a setback and a company-ending event.
Host:You keep saying "data." Why is the data the thing people forget to fight for?
Expert:Because in a decade-long program the data is the asset. The molecule without the trial results is a chemistry lesson. If your partner runs the study, controls the database, and terminates, you can own the patent and still be unable to talk to another buyer — you've got nothing to show them. I've seen a reversion clause that returned the compound but not the safety file. Useless.
Host:Let's talk incentive alignment, because that's the phrase everyone uses and nobody defines.
Expert:Alignment means the partner loses money when they slow-walk your drug. Concretely: diligence obligations with teeth. A clause that says the partner must spend a minimum amount per year and hit development steps by fixed dates, or the licence reverts. Without it, a big pharma can park your asset for three years while they prioritise their internal pipeline, and you can't do a thing.
Host:Doesn't a small biotech have zero leverage to demand that from a partner ten times its size?
Expert:Less than they think, more than they act on. The leverage is at signing, before the ink dries — that's the only moment you hold the asset outright. One number to anchor on: the royalty. A typical range is 8 to 15 percent of net sales. If you give away the high end and take a thin upfront, you've mortgaged the one revenue stream that pays out if the drug actually works.
Host:So what should a CFO trade away and what should they never trade?
Expert:Trade the upfront. Take less cash today if you must — that's recoverable. Never trade reversion, never trade the data return, never trade the diligence obligation. Those three are what keep you alive when the science goes sideways, which statistically it will more often than not.
Host:One concrete move for the CFO listening who's negotiating a term sheet this quarter.
Expert:Open your model and rebuild the deal value using real transition probabilities — that 5 to 7 percent survival rate, stage by stage. If the risk-adjusted number is a fraction of the headline, and it will be, walk back into the room and spend your remaining leverage on reversion and diligence, not on a bigger biobucks number nobody will ever pay.
Host:The press release is vanity. The reversion clause is survival.
Expert:Put that on the wall.
Host:Sources for today's episode: Accounting Today, Financial Times, CNBC Finance, CFO Dive. That's it. The CFO decision tools are live at mba-training.com.
The fundamental tension in any pharma partnership is time. A small biotech running a Phase II trial might have 18 months of runway. Its large-cap partner operates on a five-year portfolio planning cycle. Both parties sign a term sheet as if they share the same urgency, and then spend the next decade discovering they do not. The mismatch is not a negotiating failure; it is structural, and it has to be priced into the deal architecture before anyone signs.
Rising Treasury yields in 2026 have tightened the screws further. With the 10-year yield staying elevated, the risk-adjusted net present valuenet present valueNet Present Value is the sum of an investment's future cash flows discounted to today, minus the initial outlay. A positive NPV signals value creation.View full definition → of a late-stage asset compresses. Partners know this. They are using the macro environment to push upfronts down and weight more value into back-end milestones that may never be reached. A biotech CFO who accepts that logic without a fight is ceding negotiating ground that will not come back.
Building a milestone schedule that actually pays out
The first move is to mapmapUsing software to automate repetitive marketing tasks and campaigns, enabling personalisation at scale across channels like email, web, and social.View full definition → every regulatory gate to a specific payment trigger, and then add a secondary trigger for timeline slippage caused by the partner. Most term sheets list milestones like "IND filing," "end of Phase II meeting," "NDA submission," and "first commercial sale." What they omit is who controls each gate. If the partner is responsible for regulatory strategy post-licensing and they delay an NDA submission by 18 months to manage their internal portfolio, your milestone does not trigger. You have no leverage unless the contract assigns financial consequences to that delay.
Build in a "diligence clock": a rolling 12-month review that requires the partner to demonstrate that the asset is on the development plan they committed to at signing. Roche's collaborations with smaller biotechs have used this mechanism in various forms. If the asset slips from the partner's active pipelinepipelineAll active sales opportunities across the stages of the sales process, together with their combined potential value and probability of closing.View full definition → without formal notice and consent, the rights should revert or a predefined compensatory payment should vest.
Second, negotiate milestone definitions at the molecular level. "Positive Phase III results" is not a definition. Define it: the primary endpoint achieved at the pre-specified alpha level, in the ITT population, with a hazard ratio or effect size agreed in advance. AstraZeneca's 2023 collaboration structure with Daiichi Sankyo on antibody-drug conjugates set a useful precedent here: milestone payments were tied to specific regulatory designations (Breakthrough Therapy, Priority Review) in addition to trial outcomes, because those designations have economic value to the partnering company and are partially within its control.
Third, split your milestones into clinical, regulatory, and commercial tranches. Do not let a partner collapse them. Each tranche carries a different risk profile. Clinical milestones reflect your science. Regulatory milestones partly reflect the partner's relationship with the FDA or EMA and their submission quality. Commercial milestones depend on the partner's sales force, pricing strategy, and market access capability. A CFO who values the pre-revenue pipeline using risk-adjusted NPVNPVNet Present Value is the sum of an investment's future cash flows discounted to today, minus the initial outlay. A positive NPV signals value creation.View full definition → will already have these tranches separated in the model; the contract should mirror that structure.
How do you protect royalty rates when a partner has control over pricing?
The short answer: tiered royalties keyed to net sales thresholds, with a floor that survives any discounting the partner applies to win formulary access. The standard pitfall is accepting a royalty on "net sales" without defining the deductions. Partners routinely subtract co-pay assistance, managed care rebates, distribution fees, and chargebacks before calculating the royalty base. In US markets with aggressive PBM negotiations, those deductions can reduce net sales to 60-70% of gross. Negotiate a cap on allowable deductions, typically 20-25% of gross, and define it by category.
Also negotiate a minimum annual royalty during the commercial phase. If the partner under-promotes the asset, a floor payment forces a choice: either invest in the brand or return the rights.
The structural mistakes that sink milestone deals in practice
The most common failure is accepting a lump-sum upfront in exchange for reduced milestones. The logic looks sound when you need capital to fund the next trial. The problem is that upfronts are taxed as ordinary income and carry no contingency. Milestones, structured correctly, can be tied to events you influence. Swapping future optionality for present cash is rational only if your burn rateburn rateBurn rate is the speed at which a company spends its cash reserves, usually measured per month, before reaching profitability or raising more funding.View full definition → and runway analysis shows you will not survive to collect those milestones otherwise. Run that calculation explicitly, not as a gut check.
A second failure is ignoring the change-of-control clause. If your large-cap partner is acquired, the new parent may deprioritise your asset entirely. Ensure the agreement includes a change-of-control provision that either vests a portion of remaining milestones immediately or grants you a buy-back right at a pre-agreed formula. Bristol Myers Squibb's acquisition of Celgene in 2019 left several smaller partners renegotiating development commitments that the new parent had no interest in honouring. That was not a legal failure; it was a drafting failure.
The third mistake is allowing the partner to hold IP rights to combination therapies or reformulations. If your molecule is later developed as a fixed-dose combination with another partner asset, the economics of that combination may sit entirely outside your original agreement. Define the scope of licensed IP tightly and require a separate negotiation for any new indication or formulation.
Finally, watch the sublicensing terms. If the partner can sublicense to a third party in ex-US markets, your royalty on those sublicensed sales is typically a fraction of what you would earn from a direct license. Either negotiate the sublicense royalty at the time of deal signing, or retain approval rights over sublicense terms.
Start this week
- Pull your current or draft term sheet and map each milestone to its controlling party. Flag any milestone the partner controls without a diligence clock.
- Model three royalty scenarios using the gross-to-net deduction caps you have accepted (current language, 20% cap, 25% cap). Quantify the difference in NPV across a 10-year commercial period.
- Draft a change-of-control provision with your legal counsel and benchmark it against public deal disclosures from the last three years. BMS-Celgene and Pfizer-Arena Pharmaceuticals both offer instructive public filings.
- Identify which milestones in your model carry regulatory designation triggers (Breakthrough Therapy, Fast Track, PRIME in Europe) and verify that they appear in the commercial milestone schedule, not just as narrative footnotes.
Milestone-based deal structures are not inherently protective; they are only as strong as the definitions and enforcement mechanisms built into them. A CFO who treats the milestone schedule as a financial model input rather than a negotiating output will spend years watching value leak through imprecise language.
The full course on this sector:Finance in Biotech & MedTech.
Frequently asked questions
What is a reasonable upfront payment in a biotech pharma licensing deal?
Upfront payments in pharma licensing vary significantly by stage, but as a rough benchmark, deals signed at Phase II typically see upfronts of 5-15% of total deal value, with the remainder in milestones and royalties. AstraZeneca's partnership with Daiichi Sankyo, for example, included an upfront that was a fraction of the headline deal value, with the bulk tied to regulatory and commercial milestones.
Can a biotech company renegotiate milestone terms after a deal is signed?
Renegotiation is possible but uncommon without a formal trigger, such as a material change in the regulatory pathway or a protocol amendment that alters the primary endpoint. Biotech CFOs who want renegotiation rights should build them in at signing through "development plan review" clauses that require mutual agreement on any protocol change, which creates a natural window for broader commercial renegotiation.
How do royalty tiers typically work in pharma licensing agreements?
Royalty tiers in pharma deals escalate based on net sales thresholds, often stepping up from 8-10% on the first layer of sales to 15-20% above a defined revenue ceiling. The key risk is how "net sales" is defined: without a cap on allowable deductions such as rebates and chargebacks, partners can reduce the royalty base by 30-40% in markets with aggressive managed care negotiations.
What happens to milestone payments if a partnered asset fails in Phase III?
If a Phase III trial fails, milestone payments tied to that event do not vest and are generally not recoverable. Biotech CFOs should model this outcome explicitly in their risk-adjusted NPV analysis, assigning trial success probabilities by phase. Some agreements include a "data package fee" payable upon trial completion regardless of outcome, which partially compensates the biotech for the data generated.
Go deeper
The lessons that take this article further, free to read.
- 1Valuing pre-revenue pipelines with risk-adjusted NPVFinance in biotech and medtech
- 2Deal and market benchmarks: multiples, upfronts and IPO comps in the US and EuropeFinance in biotech and medtech
- 3Reading burn rate and runway like a biotech CFOFinance in biotech and medtech
- 4Earnouts, contingent consideration, and deal termsM&A, corporate development & tax
- 5The build vs. buy vs. partner decision frameworkM&A, corporate development & tax
Sources
- On the move: BDO Alliance, Abacus add member firms
- The real lesson from the Man City affair
- Debt-hungry AI companies face increased risk as bond yields spike
- Big dreams and tiny revenue are the new norm for AI IPOs
- US housing crunch puts private equity in midterm campaign crosshairs
- Burnham to revive Help to Buy scheme in bid to aid first-time buyers
- The 10-year Treasury yield is at its highest in nearly two decades. How we got here
- Medicines run short in Iran under US sanctions
- Tax Fraud Blotter: Jet skis, dirt bikes and trailers, oh my!
- IRS sees less demand for paper tax forms
- X Strategies sues co-founder for $29M embezzlement scheme
- Inflation slams consumer sentiment to four-month low: survey
- Tech news: Deloitte debuts AI-enabled M&A platform
- What accounting firms can learn from the music industry
Finished reading?
Validate your read to earn XP and feed your radar.