The Foundation Tax

Everyone in biopharma knows what a good launch needs. The playbooks are written, the causes named for twenty years. So why do launches still miss?

Because this was never a knowledge problem. It is a capacity problem. Run one launch a year and you will never notice. Run a portfolio, where introductions, indication expansions and market entries hit the same calendar and pull on the same data, content and governance teams, and the same work gets rebuilt every time: definitions, rights, assets, workflow state. That recurring bill is the foundation tax, and it lands hardest in the two places that decide readiness, data and content.

The paper makes two arguments that will annoy someone in your building. One is about what AI agents should be allowed to decide. The other is about what to build first, where the order matters more than the budget. And it does something vendor papers almost never do: it publishes the test that would prove us wrong, so you can hold us to it.