The Case Files
Ten engagements across nearly twenty years, in the order they happened. Each names the decision, the authority it was made under, and what was built.
A patient who cannot get diagnosed is a forecast that was never real.
A patient who abandons at authorization is a rebate you paid for access nobody used.
A patient who cannot afford the copay is a deduction that bought nothing.
A patient who never returns the call is a hub investment with no yield.
Every case below is one of those sentences, with a date on it.
Five rooms, one question
No performance figures appear anywhere in this set.
- The decision, and the pressure it was made under
- The authority I actually held, and what I did not control
- The architecture, governance and method I built
- What the decision cost
- What I would do differently
- Revenue, growth, margin or market share
- Forecast accuracy or variance reduction
- Any outcome I cannot separate from concurrent market movement
- Any figure without a source record, a defined denominator and a documented calculation
- The name of a single client
Former employers are named because that is professional history. Clients never are. A claim bounded by its own limits is harder to attack than one that is not.
The ten
Meeting the Patient Before the Label
What changes at approval is not the product and not the patient. It is who is looking at them, and what they are counted for.
One Method, Eight Countries
Standardizing clinical trial forecasting, and learning the sequence I have used ever since.
Which Part of What Worked
Global launch sequencing taught me humility. The specialty channel turned out to be the whole thing.
Structure Instead of Spend
A competitive entry turned a category with room in it into a red ocean. The question was whether to answer with money or with architecture.
Can We Actually Find Them
A label narrower than the trial that supported it, and the question no forecast model asks.
Building the Road to Revenue
A first commercial launch with nothing to inherit, and the governance decision that mattered more than any design choice.
Simplify, Standardize, Automate, Measure
A multi-year rebuild of a finance organization, run as a sequence rather than as a structure.
The Leak I Built
I had constructed a Silent Leak™ inside my own finance organization, and I did not see it until it was already there.
Every Function Green, the Aggregate Red
A specialty product with acceptable coverage, underperforming its forecast, and a working explanation that was wrong.
Finance Should Own the Number and Not the Operating Model
Gross-to-Net is treated as a close process because that is where it becomes visible. It is downstream of everything that created it.
Twenty Five Years, One Question
A patient is diagnosed. A treatment exists. Their physician prescribes it. And still they may not receive it.
It did not start in pharma. It started in financial operations, which is to say it started by learning how organizations actually run underneath what they say about themselves. Every number a company reports is the residue of something a person did, and if you follow the number backward far enough you always arrive at a human being.
Follow the patient far enough and you will find the economics. Follow the economics far enough and you have to ask what happened to the patient.