Biotech business development: how licensing deals actually get done
Most biotech licensing conversations fail before the term sheet. Not because the science is weak, but because founders do not understand how pharma BD actually works from the inside.
Most biotech founders treat a pharma licensing conversation like a fundraising pitch. The science, the slides, the story. But pharma BD is not a fundraise. It is a procurement process run by a committee of people who will never meet you, evaluating an asset against criteria you were not given, on a timeline you cannot control. Understanding how that process works from the inside changes everything about how you prepare for it.
Why pharma BD is not a VC conversation
A venture capital investment is fundamentally a bet on a person and a story. VCs fund conviction. They invest in founders who can execute against uncertainty, and they expect to be wrong a significant percentage of the time. The portfolio model absorbs failures.
Pharma BD operates on a completely different logic. A licensing deal is not a bet on a person. It is a procurement decision for a specific asset that fills a specific gap in a specific pipeline, evaluated against a specific set of internal criteria. The pharma company is not taking a risk on you. It is solving a problem it already knows it has, with an asset it has concluded could solve it.
This distinction matters because it changes what you need to bring to the conversation. A VC wants to believe in you. A pharma BD team wants to verify that your asset is what they need. The first question is relational. The second is technical and commercial. Founders who treat a BD meeting like a VC pitch are solving the wrong problem.
A VC invests in conviction. A pharma BD team procures a solution. Preparing the same material for both audiences is one of the most common and costly mistakes in early-stage biotech.
How pharma BD actually works internally
Most external parties never see what happens after the first BD meeting. Understanding it changes how you prepare for everything before it.
The BD contact you meet is rarely the decision-maker. Their role is to filter: to assess whether the asset is worth escalating internally. If they escalate, the asset goes to a committee, typically a cross-functional group including research, clinical, regulatory, finance, and portfolio strategy. Each function evaluates the asset against its own criteria. Research asks whether the mechanism is credible. Clinical asks whether the development path is executable. Regulatory asks whether the approval pathway is realistic. Finance models the deal economics. Portfolio strategy asks whether this fills a gap or creates a conflict.
The asset needs to survive all of these evaluations, not just one. An asset with compelling efficacy data but a weak regulatory pathway will be killed by the regulatory team regardless of what the research team thinks. An asset with a clear pathway but a deal structure that does not model well will be killed by finance.
The second thing founders rarely know is that the BD contact needs an internal champion. This is the person who will defend the asset in the committee meeting, push back when a function raises an objection, and keep the conversation alive when it would otherwise go quiet. BD contacts become champions for assets they believe in. They stop championing assets when they are given reasons not to. Every interaction you have with a BD contact is either building or eroding their willingness to go to bat for you internally.
The BD licensing funnel: four stages
The Biotech BD Licensing Funnel describes the four stages that every licensing conversation passes through, with the specific action required to advance from each stage to the next. Most deals that fail do so by stalling between stages, not by receiving an explicit rejection.
Stage 1: Signal
The first contact, typically triggered by a conference presentation, a publication, a warm introduction, or an unsolicited approach. The pharma BD contact determines whether the asset is worth a further conversation.
What advances from this stage: A clear asset summary that maps the science to a specific pipeline need, with enough data to warrant a second meeting but not so much that the BD contact is overwhelmed before they have decided whether to champion you.
Stage 2: Evaluation
The asset is under active internal assessment. Confidential disclosure agreements are signed, data packages are shared, and the BD contact is building the internal case. This stage can last weeks to months.
What advances from this stage: A data package organised around the pharma BD team's evaluation criteria, not around the scientific narrative. The internal committee will look for the same things in the same order regardless of how the data is presented. The question is whether you have made it easy or hard for them to find what they need.
Stage 3: Term sheet
The asset has cleared internal evaluation and the pharma company is prepared to negotiate. A term sheet is not an offer. It is an opening position. The structure of a biotech licensing term sheet is relatively standard: upfront payment, development milestones, commercial milestones, royalties, and definitions of the licensed field and territory.
What advances from this stage: A clear understanding of your walk-away positions before the term sheet arrives, not during the negotiation.
Stage 4: Due diligence and close
The deal is agreed in principle and the legal, regulatory, and financial teams are conducting detailed verification. This stage kills more deals than people acknowledge. Data room inconsistencies, undisclosed IP encumbrances, and team instability all surface here.
What advances from this stage: A data room that passes scrutiny without surprises. Everything in the due diligence process should confirm what was already represented, not reveal new information.
Why deals die at each stage
Most licensing conversations do not end with a formal rejection. They end with silence. The BD contact stops responding. Internal escalation is delayed indefinitely. The asset is described as not fitting current priorities. Understanding what kills a deal at each stage allows founders to prevent it.
Deals die at stage 1 because the asset summary does not make the pipeline fit immediately legible. The BD contact is evaluating dozens of inbound opportunities. An asset that requires explanation before its relevance is clear will not be championed.
Deals die at stage 2 because the data package is organised for scientific audiences rather than procurement committees. The sequence in which data is presented, the framing of each result, and the explicit connection between data and development pathway all determine whether the committee can build a case or has to make one for themselves.
Deals die at stage 3 because founders negotiate in the moment rather than from a prepared position. A term sheet negotiation moves quickly. Founders who have not worked through their acceptable ranges on upfront payments, milestone definitions, royalty rates, and field restrictions are at a structural disadvantage against BD teams that negotiate deals every quarter.
Deals die at stage 4 because due diligence reveals something the data room did not. This is not always fraud. It is often incomplete documentation of IP ownership, undisclosed third-party licences, or inconsistencies between what was represented in the BD meeting and what the legal record shows.
Most licensing deals do not fail because of the science. They fail because the asset was not presented in the language of procurement, or because due diligence found something the data room did not anticipate.
What the data package needs to do
The data package for a pharma BD conversation serves a different purpose from a scientific publication or a grant application. It needs to answer five questions in a sequence that matches the internal evaluation process of the target company.
First, does this asset address a gap in our pipeline? This is the strategic fit question. It should be answerable in the first two pages without reading the science.
Second, is the mechanism credible? This is the research function's question. It requires efficacy data in a model that is relevant to the indication and a biological rationale that is coherent.
Third, can this be developed? This is the clinical and regulatory function's question. It requires a development plan, a regulatory pathway assessment, and an honest account of the risks and how they will be addressed.
Fourth, what does the deal look like economically? This is the finance function's question. It requires a revenue model, a competitive landscape analysis, and a realistic account of what the asset will be worth at each development milestone.
Fifth, is the IP clean? This is the legal function's question. It requires a freedom to operate assessment and a clear account of what is owned, what is licensed, and from whom.
A data package that answers these five questions in this sequence, using the language the evaluation committee uses rather than the language of the scientific literature, is structurally more likely to survive internal evaluation than one that does not.
Like what you're reading?
Subscribe for more strategic notes on biotech and venture design.
Frequently asked questions
Continue with
Pharma BD is a process that runs on its own logic and its own timeline. Understanding it from the inside is not optional preparation for a licensing conversation. It is the preparation.
Book a 30-minute call