Medical Affairs — the function that translates clinical evidence for physicians, payers, and regulators — has been demonstrating its value for decades. The problem is not evidence. It is power.
McKinsey published “A Vision for Medical Affairs 2030“ and chose a metaphor of three “coequal pillars” to represent R&D, Medical Affairs, and Commercial. I have been in Medical Affairs for more than two decades, and that analogy is the most accurate depiction of the structural problem I have ever seen… I do not think they meant it that way.
Three pillars, standing apart. That is the daily experience of most Medical Affairs professionals I know.
We Have Been Here Before
For at least ten years, publications from McKinsey, MAPS, and others have told us…
… Medical Affairs is increasingly critical in elevating the understanding of disease and the treatments available;
… the function is growing in strategic importance in supporting both clinicians and patients;
… leadership finally understands the value we bring in supporting the overall company vision.
And yet, the function remains structurally undefined. It is neither fully accepted by R&D or by Commercial. Instead, companies pick the parts of Medical Affairs they find convenient — evidence generation, real-world studies, KOL engagement — and leave the rest.
The result? The value professional Medical Affairs leaders deliver does not convert into the one thing that matters inside pharma and biotech organizations: decision-making power. In my view, that is not a perception problem; it is a structural one.And it will not be solved by another industry report telling us we matter.
The Problem Is Currency
Over the past two decades, Medical Affairs has expanded the scope of its responsibilities in real terms. Thriving Medical Affairs teams took on evidence generation, real-world studies, HTA dossiers, health economics support, and patient engagement. Development stepped back from many of these, as did Commercial. Today, Medical Affairs does more than ever — without claiming the strategic seat that comes with that added responsibility.
The refrain has long been that Medical Affairs needs to better demonstrate its value. I disagree with the framing. Demonstrating value is not the same as converting value into decision power.
That disconnect is a problem. Because unless Medical Affairs operates with genuine authority — at the early pipeline stage, at endpoint design, at data strategy, at HTA planning — decisions will be made that are not recoverable.
You cannot…
… undo a Phase 3 endpoint design that ignores the real-world treatment landscape;
… reconstruct a scientific narrative that was never built;
… recapture the costs of late Medical Affairs involvement that are paid at launch, at access negotiations, and in MSL teams that inherit gaps they did not create and cannot fully close.
Third Pillar Framing Is the Wrong Model
Medical Affairs does not belong between R&D and Commercial. It belongs inside both, at the right stage, with the right mandate.
Medical Affairs is the function that moves through the lifecycle of the asset, changes shape as the asset changes stage, and hands off cleanly between phases. It is the function that understands what the end state needs to look like and works backward from it from day one.
- In early development… Medical Affairs should set the scientific narrative, align endpoints with clinical reality, and build the KOL foundation that will carry the asset through filing.
- In late development and at launch… Medical Affairs transitions to external education, guideline influence, and field execution.
- Post-launch… Medical Affairs owns real-world evidence, lifecycle management, and ongoing scientific exchange as new therapies emerge in a treatment algorithm.
This is not a supportive role — it is a different kind of leadership at each stage.
Structural Requirements
Building a full Medical Affairs organization from pipeline entry is not realistic nor necessary. Rather, what is needed is an experienced senior presence at the table — before Phase 3 recruitment closes.
This is a specific kind of role — not a Medical Director focused on study execution. What is needed is a Medical Affairs professional who has already built a launch, negotiated an HTA dossier, managed a KOL network, and knows which data gaps kill access conversations. Someone who can look at an early molecule and ask the questions that need to be answered today, not in three years.
This requires dedicated bandwidth, not episodic involvement, but sustained presence across the full arc of a program. The value is front-loaded so that decisions made at this stage hold. A small senior specialist team deploys at pipeline entry, addresses a defined set of lifecycle-critical gaps — endpoint alignment, scientific narrative, KOL foundation, evidence package design — and then transitions cleanly into the traditional launch-stage build as the program scales.
Each phase requires different skills and different headcount. The specialist team that does the early work is not the same team that runs the launch. This small senior team needs to influence decisions across multiple programs.
Finally, structure without infrastructure fails. Organizations that get this right build systems where early Medical Affairs decisions are documented and visible, so the scientific narrative built in Phase 1 is still driving decisions at launch.
The Conversation Begins with Medical Leaders
External perspectives are informative. But if the argument for Medical Affairs’ strategic role is always made by external consultants and strategy firms, it will always be received as external advice, not an internal mandate.
Medical leaders need to make this case from inside their organizations and in public. We need to define what early Medical Affairs looks like structurally and commit to the infrastructure that enables it — how it is resourced, what it is accountable for, and how it is measured. Not as a response to the next industry survey or reorg, but as a clear position that medical leaders own.
The pattern is consistent. The earlier Medical Affairs is genuinely integrated, the fewer problems at launch. The later it arrives, the more it inherits gaps it did not create and cannot fully close.
