One Method, Eight Countries

by | Sep 1, 2026 | Gross-to-Net

For life sciences teams

Standardizing clinical trial forecasting, and learning the sequence I have used ever since.

I owned the methodology but not the country organizations, so one accrual and forecasting method had to earn adoption across eight countries on quality rather than mandate. Standardize before you automate. A process automated before it is standardized just runs the wrong way faster.

Bayer · R&D FinanceArchitectureThe 5 Layers of GTN™ Layer 1

Decision context

Clinical trial forecasting was fragmented. Country-level accruals were inconsistent with one another, and finance had no direct visibility into how the contract research organizations were executing.

One forecasting method earning adoption across eight countries Clinical trial forecasting was fragmented, with country-level accruals inconsistent with one another and no direct finance visibility into how the contract research organizations were executing. The problem was not poor work. Several groups were each doing the work reasonably, in several different ways, and nobody could reconcile the results, which is harder to fix than incompetence because there is nobody to correct. Owning the method but not the country organizations meant the method had to earn adoption on quality rather than mandate. The rule that came out of it is to standardize before automating, because a process automated before it is standardized simply runs the wrong way faster. THE PROBLEMSeveral groups,several methodsEach doing it reasonably.Nobody could reconcile.THE CONSTRAINTI owned themethod, not thecountry organizationsthat had to run it.SO IT HAD TOEarn adoptionOn quality ratherthan on mandate.THE RULE LEARNEDStandardizebefore automatingOr the wrong way justruns faster.That is a harder problem than incompetence, because there is no one to correct.I built a method and then sold it. The version where they build it with me is slower to produce and faster to adopt.
FIGURE 1. Owning the method and not the organization changes what the method has to be good enough to do.

The problem was not that anyone was doing the work badly. Several groups were each doing it reasonably, in several different ways, and nobody could reconcile the results. That is a harder problem than incompetence, because there is no one to correct.

My authority

I owned the method. I did not own the country organizations that had to adopt it, and I did not own the contracts with the research organizations. Anything I built had to be good enough that people chose it, because I could not require it.

What I built

  • A single accrual and forecasting method for US-based clinical trials, documented well enough to be applied the same way by people who had not been in the room when it was designed.
  • Data tracking that made execution visible to finance directly, rather than inferred from invoices after the fact.
  • Internal controls, developed, implemented and validated, so the method would survive the people who built it. This is the part that gets skipped and it is the part that determines whether anything lasts.

The trade-off

Standardization costs local judgment. A country team that had built a good method for its own conditions had to give it up for one that worked acceptably everywhere. That is a real loss and pretending otherwise is how you lose the room.

What I would do differently

I would have brought the country teams into the design rather than into the rollout. I built a method and then sold it. The version where they build it with me is slower to produce and faster to adopt, and I did not understand that yet.

Standardize before you automate. A process automated before it is standardized just runs the wrong way faster, with more confidence than you had before.

Former employers and products referenced are matters of professional history. The views expressed are my own and are not made on behalf of any current or former employer, and no commercial outcome for any named product is attributed to me personally. All trademarks belong to their respective owners.

What the method still cannot see is in You Can Be Exactly Right About the Wrong Number. The ten places a patient can leave are on the philosophy page, and terms used here are defined in the glossary. The patient-side view of the same ten exits is at Where did your treatment stop?

Free by email

The 5 Layers of GTN™, in your inbox

Gross-to-Net from someone who has run it. Tuesdays and Thursdays.

Subscribe on Substack
Disclosure: who pays for this work
LinkedIn Newsletter Podcast YouTube Instagram