We Have More Data Than Ever. I'm Not Sure We Have More Insight.

A physician friend of mine recently got a call from a pharmaceutical company. Two members of their team wanted to know if they could come to her office and spend half a day watching her practice.

I smiled when she told me. Somewhere in that company, someone has figured out what the industry quietly gave up.

For most of my career as a marketing researcher, I did in-office interviews. Hundreds of them. I'd fly into a city, sit in the waiting room, and by the time the physician walked in to meet me, I already knew a great deal about them.

I knew how many chairs were in the waiting room and whether they were full. I knew how many staff they had and how they spoke to patients. I knew the condition of the office, the neighborhood it sat in, the diplomas on the wall, the photographs on their desk. I knew which pharmaceutical companies had been in recently, because their tchotchkes were on the shelves and their promotional material was in the exam rooms. I knew what they read, what they valued, and often what kind of practice they were trying to build.

None of that shows up in a transcript. None of it shows up on Zoom. And almost none of it shows up when a physician drives to a research facility, sits in a conference room, and answers questions for an hour.

Today, most physician research is done in facilities or over video. The economics are obvious — more interviews per day, lower travel costs, broader geographic reach. And there are real gains in scale. But there are quiet losses too, and I don't think the industry has fully reckoned with them.

The first loss is selection. The physicians who agree to drive to a facility on a Tuesday afternoon are, by definition, not the busiest ones. The highest-volume physicians — the ones whose behavior actually moves markets — are the hardest to recruit and the least likely to leave their offices. When you interview in their offices, you get them. When you interview in facilities, you get whoever had time.

The second loss is context. A physician's office is an ecosystem. Meeting them inside it tells you things they would never think to tell you in an interview, because to them it's just how the world works. The stack of samples on the counter, the layout of the exam rooms, the pace of the front desk — all of it is data. When I was Marketing Research Director at a division of Becton Dickinson, I did in-lab and hospital interviews constantly, and I came to understand not just the physicians but the labs, the workflows, and the systems they worked inside. That understanding didn't come from questions. It came from being there.

The third loss is body language, and specifically the small tells that only surface when someone is on their own turf. A physician on Zoom is performing an interview. A physician in their office, between patients, in their own chair, is being themself.

Over time, those hundreds of in-office interviews became the foundation of everything I understood about physicians as customers. It's what allowed me, at Becton Dickinson, to sit in leadership meetings not just as the voice of the customer — anyone with a survey can claim that — but as someone who could speak for them because I'd spent so much time inside their world. It's the foundation that ultimately shaped the RAMPx model I later developed for assessing the commercial potential of new products.

I'm not arguing we should go back to the old way. The economics don't support it and the technology has moved on. But I do think we've drifted into a place where we mistake volume for understanding. We run more interviews, generate more data, build more decks — and we may know our customers less well than we did thirty years ago.

The pharmaceutical company that asked to spend half a day in my friend's office understands this. They know that one afternoon inside a real practice will teach them things a hundred facility interviews will not.

Sometimes the most valuable research isn't the study you commission. It's the chair you sit in, in the room where the work actually happens.