Biotech strategy is translation work. Not translation between languages, but between epistemic regimes — each with its own standards of evidence, its own tolerance for uncertainty, and its own definition of success.
Four languages
A biotech strategist must be fluent in at least four languages:
- Scientific truth — what the data actually shows, with appropriate confidence intervals and caveats
- Clinical usefulness — what would change a physician's decision or a patient's outcome
- Operational feasibility — what can actually be manufactured, distributed, and administered
- Economic reality — what payers will reimburse, what prices the market will bear
These languages are not mutually translatable. A scientifically true finding may be clinically useless. A clinically useful intervention may be economically unviable. An economically attractive market may be scientifically empty.
The strategist's job
The strategist's job is not to pick one language and ignore the others. It is to hold all four in tension and make decisions that are honest about the tradeoffs.
This requires a specific kind of intellectual honesty that is rare in both science and business:
- Scientists who can acknowledge when their findings are not yet clinically actionable
- Clinicians who can acknowledge when standard of care is not evidence-based
- Business leaders who can acknowledge when market opportunity does not equal patient value
Common failure modes
Science-first blindness: Pursuing technically interesting questions with no path to clinical or commercial impact.
Market-first blindness: Chasing large markets with scientifically weak differentiation.
Clinical-first blindness: Optimizing for physician adoption without considering whether the intervention actually improves outcomes.
Operational blindness: Designing beautiful science that cannot be manufactured at scale.
A framework for translation
Before any strategic decision, ask:
- What is the strongest scientific claim we can make honestly?
- If true, who benefits and how measurably?
- What would it take to deliver this benefit at scale?
- Who pays, and why?
If any answer is "we don't know yet," that is the most important finding — not a problem to paper over.